An intelligent fence monitoring and management system for animal husbandry

Abstract

The present invention provides an intelligent fence monitoring and management system for the livestock industry, specifically relating to the technical field of electrical digital data processing. It includes a video acquisition module for acquiring monitoring videos corresponding to at least one monitoring area; a region analysis module for analyzing each monitoring area based on the monitoring videos to obtain monitoring objects that appear in the warning area; a monitoring data acquisition module for acquiring the motion information and environmental information of the monitoring objects; a first calculation module for calculating the motion trend degree and environmental trend degree respectively according to the motion information and environmental information; a second calculation module for calculating the fence crossing possibility based on the motion trend degree and environmental trend degree; and a warning sending module for determining whether to send warning information about the monitoring objects to the terminal based on the fence crossing possibility. The present invention combines the environment and the behavior of the monitoring objects to monitor the objects in the warning area in real time, effectively improving the monitoring efficiency of the livestock industry and enhancing the management efficiency.

Description

Technical Field

[0001] The present invention relates to the technical field of electric digital data processing, and particularly to an intelligent fence monitoring and management system for animal husbandry. Background Art

[0002] In the existing animal husbandry management, in order to prevent livestock from climbing over the fence, real-time status monitoring can be carried out, and computer vision technology is used to analyze the movement status of livestock to determine whether there is a risk of livestock climbing over the fence. However, traditional animal husbandry fences are generally made of metal materials such as iron wire mesh and steel pipes. Such fences are exposed to harsh environments such as wind, rain, and livestock biting for a long time, and are extremely prone to problems such as rusting and breaking, and structural deformation. Therefore, if only the movement status of livestock in the image is analyzed without considering the fence environmental status, there will still be a serious risk of livestock escaping. Summary of the Invention

[0003] In order to solve the above technical problems, the purpose of the present invention is to provide an intelligent fence monitoring and management system for animal husbandry. To achieve the above purpose, the technical solutions adopted by the present invention are as follows:

[0004] The present application provides an intelligent fence monitoring and management system for animal husbandry, including: a video acquisition module for acquiring monitoring videos corresponding to at least one monitoring area, each of the monitoring areas including a section of fence and a warning area adjacent to the fence; a region analysis module for analyzing each of the monitoring areas based on the monitoring videos to obtain monitoring objects that appear in the warning area; a monitoring data acquisition module for acquiring movement information and environmental information of the monitoring objects, the movement information including speed and movement direction within a preset time period, the end moment of the preset time period being the current moment, and the environmental information being fence parameters within the monitoring area where the monitoring objects are located; a first calculation module for calculating a movement trend degree and an environmental trend degree respectively according to the movement information and the environmental information;

[0005] a second calculation module for calculating a fence crossing possibility based on the movement trend degree and the environmental trend degree; and a warning sending module for determining whether to send a warning message about the monitoring object to the terminal based on the fence crossing possibility.

[0006] In a possible implementation, the method for calculating the degree of motion trend includes: calculating a moving angle according to the fence parameter and the moving direction at the current moment, where the moving angle is the angle between the moving direction and the normal vector of the fence, and the normal vector of the fence is the normal vector at the fence point closest to the monitored object; taking the fence extension direction as a reference and combining the moving angle, projecting and decomposing the speeds corresponding to the current moment and the previous moment respectively to obtain the fence speeds projected on the fence extension direction, where the previous moment is the starting moment of a preset time period; calculating the degree of motion trend based on the moving angle and the fence speed components corresponding to the current moment and the adjacent moment.

[0007] In a possible implementation, the method for calculating the degree of motion trend further includes: obtaining the historical data of the monitored object, where the historical data includes the number of times of historical fence crossing and the height of historical fence crossing; correcting the degree of motion trend based on the historical data to obtain the corrected degree of motion trend.

[0008] In a possible implementation, the environmental information further includes the shortest distance from the monitored object to the fence at the current moment and the fence height within the warning area where the monitored object is located at the current moment. The method for calculating the environmental trend degree includes: analyzing to obtain an adjacent area based on the motion information, where the adjacent area is the next warning area that the monitored object enters according to the moving direction at the current moment; obtaining the fence height of the adjacent area to get the adjacent fence height; analyzing the target frame to obtain the damaged area of the fence, where the target frame is the frame corresponding to the monitored video at the target moment, and the target moment is the moment when the monitored object first appears; calculating the environmental trend degree based on the environmental information, the adjacent fence height, and the damaged area of the fence.

[0009] In a possible implementation, the second calculation module includes: a data update module for continuously monitoring the monitored object within the warning area, recording the monitoring duration and updating the degree of motion trend at each moment; a correction module for performing weighted summation on the degree of motion trend and the environmental trend degree according to the monitoring duration to obtain the possibility of fence crossing.

[0010] In a possible implementation, the data update module includes: a data secondary acquisition module for re-acquiring the motion information of the monitored object and calculating the second-order degree of motion trend; a judgment module for judging whether the second-order degree of motion trend is the same as the degree of motion trend, and if not, updating the degree of motion trend to the second-order degree of motion trend.

[0011] In a possible implementation, the early warning sending module includes: a logic module, configured to generate an early warning message when the motion trend degree is greater than a preset judgment threshold; a response module, configured to respond to the generation result of the early warning message and send the early warning message to a terminal.

[0012] In a possible implementation, the preset judgment threshold is 0.8.

[0013] In a possible implementation, the speed is collected by an acceleration sensor worn on the monitored object.

[0014] In a possible implementation, the motion direction is collected by a gyroscope worn on the monitored object.

[0015] The present invention has the following beneficial effects:

[0016] In the present invention, by combining the content of the environmental dimension and the behavior dimension of the monitored object itself, the behavior and position of the monitored object in the early warning area are monitored in real time, which can effectively improve the monitoring efficiency of the livestock industry. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0018] Figure 1 FIG. is a schematic structural diagram of an intelligent fence monitoring and management system for livestock industry provided by Embodiment 1 of the present invention;

[0019] Figure 2 FIG. is a schematic flowchart of a calculation method for the motion trend degree in Embodiment 1 of the present invention;

[0020] Figure 3 FIG. is a schematic flowchart of a calculation method for the environmental trend degree in Embodiment 1 of the present invention;

[0021] Figure 4 FIG. is a schematic structural diagram of the early warning sending module in Embodiment 1 of the present invention;

[0022] Figure 5 FIG. is a schematic structural diagram of the second calculation module in Embodiment 2 of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0023] In order to further elaborate on the technical means and effects adopted by the present invention to achieve the intended invention purpose, the following, in combination with the accompanying drawings and preferred embodiments, details the specific implementation manner, structure, features and effects of an intelligent fence monitoring and management system for animal husbandry proposed according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. In addition, the specific features, structures or characteristics in one or more embodiments can be combined in any suitable form.

[0024] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those of ordinary skill in the technical field to which the present invention belongs.

[0025] Embodiment 1:

[0026] The following specifically describes the specific solution of an intelligent fence monitoring and management system for animal husbandry provided by the present invention in combination with the accompanying drawings.

[0027] Please refer to Figure 1 , which shows a schematic structural diagram of an intelligent fence monitoring and management system for animal husbandry provided by an embodiment of the present invention. In this embodiment, the intelligent fence monitoring and management system for animal husbandry includes:

[0028] A video acquisition module, configured to acquire monitoring videos corresponding to at least one monitoring area, and each of the monitoring areas includes a section of fence and a warning area adjacent to the fence.

[0029] It should be noted that in this embodiment, the monitoring videos are collected by multiple cameras with 120-degree wide-angle lenses installed in the ranch. These cameras are all treated with waterproof and dustproof to maintain long-term stable operation, and the installation height is recommended to be 2.5 - 3 meters above the fence top, which can not only avoid animal collisions but also obtain the best monitoring perspective. At the same time, it can be understood that in this embodiment, each camera corresponds to collecting the monitoring video of a monitoring area, and there is a 5% - 10% overlapping area between the coverage ranges of adjacent cameras to ensure no dead angle in monitoring. After the cameras are installed at a position higher than the fence, the 120-degree wide-angle lenses can completely cover a monitoring area with a depth of 8 - 10 meters, so that in the shooting area of each camera, not only a section of fence is included, but also the ranges of 3 - 5 meters inside and outside the fence can be monitored synchronously. Through the multi-lens video stitching algorithm and the geographical information system (GIS) coordinate mapping, the digital stitching of multiple shooting areas can achieve continuous video coverage of all fence sections.

[0030] Specifically, in this embodiment, the warning area is a strip area range that expands 0.5 meters into the pasture with the fence as the boundary main body. Those skilled in the art should understand that during the implementation process, other values can be set according to factors such as the breed differences of the monitored objects (such as sheep and cattle), and the fence materials (electric fence / wooden fence). In this embodiment, it is recommended that 0.3 - 0.7 meters is more appropriate. Therefore, based on the above division results, in this embodiment, the possibility of the monitored object crossing the fence can be confirmed according to the area where the monitored object is located.

[0031] The area analysis module is used to analyze each of the monitoring areas based on the monitoring video to obtain the monitored objects that appear in the warning area.

[0032] It should be noted that the monitored objects mentioned in this embodiment can be animals such as sheep, beef cattle, and horses raised in the pasture. In this module, the system uses the target recognition algorithm to real-time identify the monitored objects from the video, and its technical implementation relies on the video analysis architecture deployed in the edge computing unit or the video analysis architecture of the central processor located at the administrator's residence. Regarding the target recognition algorithm, those skilled in the art can use the YOLOv5 (You Only Look Once version 5) or Faster R-CNN (Region-based Convolutional Neural Networks) model based on the convolutional neural network. Among them, the YOLO series algorithms are more suitable for real-time monitoring scenarios due to their single-stage detection characteristics, while Faster R-CNN has an advantage in terms of target localization accuracy. It should be added that regarding the model training data, this embodiment recommends that it should include animal image samples under different lighting conditions (morning haze, noon strong light, dusk backlight), and data enhancement processing (random rotation, noise injection, contrast adjustment) should be performed to improve the robustness of the model. Among them, the usage methods of YOLO or Faster R-CNN are prior arts and will not be elaborated in this embodiment.

[0033] The monitoring data acquisition module is used to acquire the motion information and environmental information of the monitored object. The motion information includes the speed and motion direction within a preset time period, the end moment of the preset time period is the current moment, and the environmental information is the fence parameters within the monitoring area where the monitored object is located.

[0034] First, the speed and direction of motion mentioned in this embodiment can be obtained by the cooperation of a three-axis speed sensor and a gyroscope. Specifically, this sensor combination is integrated inside a waterproof electronic collar. The sensor combination is integrated inside the waterproof electronic collar, and the electronic collar is worn by the monitored object and fixed by an anti-detachment buckle. Regarding the way of transmitting motion information to the central processor located at the administrator's residence, a hierarchical transmission architecture is adopted: through a low-power Internet of Things communication module built into the electronic collar, it is transmitted to the receiving module installed on the fence post in an encrypted manner of 128-bit AES through Bluetooth 5.0, zeebig communication protocol or LoRa communication protocol, and then the receiving module is transmitted to the central processor located at the administrator's residence through a fiber optic network or wireless communication for multi-threaded parallel processing.

[0035] Secondly, the duration of the preset time period mentioned in this embodiment is set to be synchronized with the timing wake-up period of the Internet of Things communication module, and can be specifically configured as a communication interval of 30 seconds / time. Through this setting, a time alignment window is formed between the data acquisition timestamp and the transmission cycle, which can not only reduce the power consumption of the wireless module, but also ensure the consistency of the motion trajectory analysis calculation timing.

[0036] In addition, regarding the fence parameters in the environmental information mentioned in the embodiment, the system queries the fence parameters corresponding to the monitoring of each camera in the preset database through the unique identification code corresponding to each camera. The preset database adopts a time-series database architecture and contains an associated data set of the unique identification code and the fence parameter table (geographical coordinates, height). At the same time, to ensure the unity of subsequent spatial calculations, in this embodiment, the reference direction of the gyroscope is the same as the reference direction of the geographical coordinates describing the fence, so as to establish a unified azimuth coordinate system.

[0037] The first calculation module is used to calculate the motion trend degree and the environmental trend degree respectively according to the motion information and the environmental information.

[0038] In this embodiment, considering the position of the fence relative to the entire warning area, it is located at the boundary part of the area. By analyzing the moving speed and direction of the monitored object, the possibility of its crossing the fence trend can be confirmed. Specifically, if the moving direction of the monitored object continuously points to the boundary direction, then the possibility of the monitored object crossing the fence trend will increase. In addition, when the moving direction of the monitored object is highly consistent with the boundary extension direction of the fence, it can be inferred that the monitored object may just be observing the surrounding environment continuously, and the possibility of its crossing the fence trend is relatively low. At the same time, the height and damage condition of the fence itself will also affect the monitored object and induce it to have a trend of crossing the fence. Therefore, in this embodiment, based on these factors, the motion trend degree and the environmental trend degree are calculated to more accurately measure the behavior of the monitored object.

[0039] The calculation method of the motion trend degree can be referred to Figure 2 . The figure shows that the calculation method of the motion trend degree includes steps S1 - S3

[0040] S1. Calculate the moving angle according to the fence parameter and the motion direction at the current moment. The moving angle is the angle between the motion direction and the normal vector of the fence. The normal vector of the fence is the normal vector at the fence point closest to the monitored object

[0041] S2. Based on the fence extension direction and in combination with the moving angle, project and decompose the speeds corresponding to the current moment and the previous moment respectively to obtain the fence speeds projected on the fence extension direction. The previous moment is the starting moment of the preset time period

[0042] S3. Calculate the motion trend degree based on the moving angle, the fence speed components corresponding to the current moment and the adjacent moment

[0043] Specifically, the calculation functional formula of the motion trend degree is as follows

[0044] ;

[0045] Wherein represents the motion trend degree of the th monitored object represents the absolute value calculation function represents the th monitored object at the th moment of the moving angle represents the th monitored object at the th moment of the speed projected onto the fence direction of the fence speed represents the th monitored object at the th moment of the speed projected onto the fence direction of the fence speed represents the th monitored object at the th moment to the shortest distance of the fence represents a non-zero constant. In this embodiment .

[0046] In the above calculation functional formula, the th moment represents the previous moment, and the th moment represents the current moment. At the same time, regarding can be obtained from the In the surveillance video at a certain moment, it is identified through a target recognition algorithm, and the real distance is obtained by calculating the relevant distance of the image; it can also be obtained by setting a positioning module in the electronic collar, and then obtaining the geographical coordinates of the monitored object through the positioning module, and calculating them in combination with the geographical coordinates of the fence. The specific implementation methods are all existing technologies and will not be elaborated in this embodiment.

[0047] As described above, when the moving speed of the monitored object increases within a preset time period and the moving direction of the monitored object is also towards the fence, it indicates that the possibility of it crossing the fence is relatively high. Specifically, that is, the angle between the moving direction of the monitored object and the normal vector of the fence can reflect whether the animal is moving towards the fence. The smaller the angle, the more likely the monitored object is to move directly towards the fence, and the greater the possibility that the monitored object will cross the fence. When the angle is close to 90 degrees (when the animal is walking almost straight along the fence boundary), it may indicate that the monitored object is constantly observing the surrounding environment, and the possibility of its crossing the fence trend is relatively low.

[0048] At the same time, in this embodiment, it is also considered that the historical behavior and data will affect the possibility of the monitored object crossing the fence. If the monitored object has tried to climb over the fence many times before or has a successful experience of climbing over the fence in the historical record, the possibility of the monitored object crossing the fence trend will be greater. Therefore, in this embodiment, it also includes correcting the motion trend degree according to the historical behavior. See steps S4 and S5 for details.

[0049] S4. Obtain the historical data of the monitored object, where the historical data includes the number of historical fence crossings and the historical jumping height.

[0050] S5. Correct the motion trend degree based on the historical data to obtain the corrected motion trend degree.

[0051] Specifically, the correction functional formula of the motion trend degree is as follows:

[0052] ;

[0053] Among them, represents the corrected motion trend degree of the th monitored object; represents the number of historical fence crossings of the th monitored object; represents the highest historical jumping height of the th monitored object, which can be obtained by analyzing the historical surveillance video; represents the fence height within the monitoring area where the th monitored object is located at the current moment; represents the The motion trend degree of a monitored object before correction.

[0054] In the above correction functional formula, represents the th number of times a monitored object has climbed over a fence in history. The more times, the stronger the willingness of the monitored object to climb over the fence, and the greater the possibility of the monitored object crossing the fence. represents the th maximum height that a monitored object can jump. When it is greater than the fence height and the difference is larger, it means that the th monitored object is more likely to successfully cross the fence. Therefore, in this embodiment, by combining and , the motion trend degree of the th monitored object crossing the fence is corrected, improving the accuracy of calculating the possibility of the th monitored object crossing the fence subsequently.

[0055] As described above, in this embodiment, it is also considered that the current condition of the fence will affect the possibility of a monitored object crossing the fence. If there are potential climbing opportunities in the current environment, such as fence damage, height, etc., the greater the possibility of the monitored object having the behavior of climbing over the fence. Therefore, it is necessary to calculate the environmental trend degree in combination with the fence height and damage situation. Specifically, for the calculation method of the environmental trend degree, reference can be made to Figure 3 , and the figure shows that the calculation method of the environmental trend degree includes Step 1 - Step 3.

[0056] Step 1: Analyze the adjacent area based on the motion information. The adjacent area is the next warning area that the monitored object enters according to the motion direction at the current moment.

[0057] Step 2: Obtain the fence height of the adjacent area to get the adjacent fence height.

[0058] Step 3: Analyze the target frame to obtain the damaged area of the fence. The target frame is the corresponding frame of the monitoring video at the target moment, and the target moment is the moment when the monitored object first appears.

[0059] Among them, the calculation method of the damaged area of the fence can be based on the analysis of the fence surface image of the target frame. After using image segmentation technology to extract the effective area of the fence, the integrity of the fence structure is identified through an edge detection algorithm. Specifically, it includes: converting the target frame of the surveillance video into the HSV color space, and segmenting the main area of the fence by using the preset color threshold of the fence material; detecting the edge contour of the fence by using the Canny operator, and identifying the grid structure of the fence in combination with the Hough line transformation; marking the abnormal areas with fractures, depressions or material deficiencies as damaged areas, and finally obtaining the numerical value of the damaged area of the fence by calculating the ratio of the pixel area of the damaged area to the total pixel area of the fence and combining the physical size parameters corresponding to the unit pixel calibrated in advance. At the same time, for those skilled in the art, an end-to-end damage recognition method for the fence surface can also be adopted by using a semantic segmentation model based on deep learning (such as the U-Net architecture). This is prior art, and the process will not be elaborated in this implementation.

[0060] Step4. Calculate the environmental trend degree based on the environmental information, the height of the adjacent fence, and the damaged area of the fence.

[0061] Specifically, the calculation functional formula of the environmental trend degree is as follows:

[0062] ;

[0063] Among them, represents the environmental trend degree of the th monitoring object; represents the maximum-minimum normalization function; represents the damaged area of the fence in the monitoring area where the th monitoring object is located; represents the total area of the fence in the monitoring area where the monitoring object is located; The th monitoring object is at the time, and the height of the fence in the monitoring area where it is located; represents the height of the adjacent fence corresponding to the th monitoring object at the time; represents the shortest distance from the th monitoring object to the fence at the time; represents a non-zero constant. In this embodiment, .

[0064] In the above calculation functional formula, The greater the difference between the two, the greater the possibility that the current th monitoring object crosses the fence; Indicates the damaged area ratio of the fence. The higher this ratio, the easier it is for the monitored object to cross the fence, and thus the greater the likelihood of the current th monitored object having a tendency to cross the fence; Indicates the th monitored object's shortest distance to the fence at the th moment. The smaller this distance, the greater the likelihood of the current th monitored object having a tendency to cross the fence. Therefore, the above calculation formula can better describe the influence of environmental factors on the fence-crossing behavior of the monitored object.

[0065] A second calculation module, configured to calculate the fence-crossing possibility based on the motion tendency degree and the environmental tendency degree.

[0066] Specifically, the calculation functional formula for the fence-crossing possibility is as follows:

[0067] ;

[0068] Wherein, Indicates the fence-crossing possibility of the th monitored object; Indicates the max-min normalization function; Indicates the corrected motion tendency degree of the th monitored object; Indicates the environmental tendency degree of the th monitored object; Indicates the first weight coefficient; Indicates the second weight coefficient. In this embodiment , for those skilled in the art, other weight coefficients can also be selected, and no specific limitation is made in this embodiment.

[0069] In the above calculation functional formula, by assigning weights to the fence-crossing possibility and the environmental tendency degree respectively and performing weighted summation, an evaluation of the fence-crossing possibility in terms of both the environmental dimension and the behavior dimension of the monitored object itself is obtained, which can better describe whether the monitored object has the possibility of crossing the fence.

[0070] An early warning sending module, configured to determine whether to send early warning information about the monitored object to the terminal based on the fence-crossing possibility.

[0071] Specifically, referring to Figure 4 , the figure shows that this module further includes a logic module and a response module.

[0072] In this embodiment, the logic module is configured to generate early warning information when the motion tendency degree is greater than a preset judgment threshold.

[0073] Among them, the preset judgment threshold is 0.8. For those skilled in the art, this preset judgment threshold can also be selected as other values according to the actual situation, and no specific limitation is made in this embodiment.

[0074] It should also be noted that in this embodiment, there is a preferred hierarchical early warning mechanism. When the motion trend degree is in the range of 0.8 - 0.9, a yellow early warning is triggered, and a real-time monitoring video including the timestamp, the geographical coordinates of the dangerous monitoring object, and the monitoring area where the dangerous monitoring object is located is generated. Among them, the dangerous monitoring object refers to the monitoring object when the motion trend degree is greater than the preset judgment threshold, and the geographical coordinates are collected by the positioning module set in the electronic collar. When the motion trend degree is in the range of 0.89 - 1.0, a red early warning is triggered, and similarly, a real-time monitoring video including the timestamp, the geographical coordinates of the dangerous monitoring object, and the monitoring area where the dangerous monitoring object is located is generated. But at the same time, an audible and visual alarm device should be activated at the fence, and a three-level gradient alarm mode is formed by combining high-frequency beeping and red light flashing to drive the monitoring object away from the fence.

[0075] The response module is used to respond to the generation result of the early warning information and send the early warning information to the terminal.

[0076] In this embodiment, by combining the content of the environmental dimension and the behavior dimension of the monitoring object itself, the behavior and position of the real-time monitoring object can be effectively improved, the monitoring efficiency of the animal husbandry industry can be effectively improved, the frequency of manual inspections can be reduced, false alarms and human errors can be reduced, and the management efficiency can be improved.

[0077] Embodiment 2:

[0078] This embodiment provides an intelligent fence monitoring and management system for animal husbandry. The difference between this embodiment and Embodiment 1 lies in the different execution contents of the second calculation module. Specifically, refer to Figure 5 , and the figure shows that the second calculation module includes:

[0079] The data update module is used to continuously monitor the monitoring object in the early warning area, record the monitoring duration, and update the motion trend degree at each moment.

[0080] The correction module is used to perform a weighted sum of the motion trend degree and the environmental trend degree according to the monitoring duration to obtain the fence crossing possibility.

[0081] Specifically, the calculation functional formula of the fence crossing possibility is as follows:

[0082] ;

[0083] Among them, represents the fence crossing possibility of the th monitoring object; denotes the Sigmoid function; denotes the modified motion trend degree of the th monitored object; denotes the environmental trend degree of the th monitored object;

[0084] In this embodiment, considering that if the current th monitored object stays in the warning area for a longer time, it indicates that the influence brought by environmental factors is smaller than that of the motion behavior. That is, when the staying time is longer, the weight ratio of the motion trend degree of the th monitored object should be larger, while the weight ratio of the environmental trend degree should be smaller. At the same time, if the staying time is too long, it should not have an infinite increasing trend. Therefore, the Sigmoid function is used to limit the positive infinite increasing influence brought by the staying time. Similarly, if the current th monitored object stays in the warning area for a shorter time, and at this time, if the fence is more severely damaged, the possibility of the current th animal crossing the fence should be greater. Therefore, the above calculation formula can better describe the above situation.

[0085] It should be noted that: the above sequence of the embodiments of the present invention is only for description and does not represent the advantages or disadvantages of the embodiments. The processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0086] Each embodiment in this specification is described in a progressive manner. The same or similar parts among the embodiments can be referred to each other, and the key points of each embodiment are the differences from other embodiments.

Claims

1. An intelligent fence monitoring and management system for animal husbandry, characterized in that, include: A video acquisition module, used to acquire a surveillance video corresponding to at least one surveillance area, each of which includes a fence and an early warning area adjacent to the fence; A regional analysis module, used to analyze each monitoring area based on the monitoring video to obtain monitoring objects appearing in the warning area, wherein the monitoring objects are animals raised in the pasture; A monitoring data acquisition module, used to acquire motion information and environmental information of the monitored object, wherein the motion information includes speed and motion direction within a preset time period, the end time of the preset time period is the current time, and the environmental information is fence parameters within the monitoring area where the monitored object is located, and the fence parameters include geographic coordinates and height; A first calculation module is used to calculate the moving angle according to the fence parameters and the movement direction at the current moment, and calculate the movement trend degree based on the moving angle and the fence speed components corresponding to the current moment and the adjacent moments, wherein the moving angle is the angle between the movement direction and the fence normal vector, and the fence normal vector is the normal vector at the fence point closest to the monitored object. The fence speed components corresponding to the current moment and the previous moment are based on the extension direction of the fence, and are obtained by projecting and decomposing the speeds corresponding to the current moment and the previous moment in combination with the moving angle; Analyzing the movement area trend based on the movement information, and obtaining the environmental trend degree in combination with the fence parameters; A second calculation module, configured to calculate the fence crossing possibility based on the movement trend degree and the environment trend degree; An early warning sending module is used to determine whether to send early warning information about the monitored object to the terminal based on the possibility of crossing the fence; The environmental information also includes the shortest distance from the monitored object to the fence at the current moment, and the height of the fence in the warning area where the monitored object is located at the current moment. The calculation method of the environmental trend degree includes: Obtaining an adjacent area based on the motion information analysis, where the adjacent area is the next warning area that the monitored object enters according to the motion direction at the current moment; Obtaining the fence height of the adjacent area to obtain the adjacent fence height; Analyze the target frame to obtain the fence damage area, wherein the target frame is the frame corresponding to the monitoring video at the target time, and the target time is the time when the monitoring object first appears; The environmental trend degree is calculated based on the environmental information, the height of the adjacent fence, and the damaged area of the fence.

2. The intelligent fence monitoring and management system for animal husbandry according to claim 1, characterized in that, The calculation method of the movement trend degree also includes: Acquire historical data of the monitored object, the historical data including the number of times the fence was climbed over and the height of the fence climbed over; The movement trend degree is corrected based on the historical data to obtain a corrected movement trend degree.

3. The intelligent fence monitoring and management system for animal husbandry according to claim 1, characterized in that The second calculation module includes: A data updating module, used to continuously monitor the monitored object in the warning area, record the monitoring duration and update the movement trend at each moment; The correction module is used to perform weighted summation of the movement trend degree and the environment trend degree according to the monitoring duration to obtain the possibility of fence crossing.

4. The intelligent fence monitoring and management system for animal husbandry according to claim 3, wherein The data updating module comprises: The data secondary acquisition module is used to re-acquire the motion information of the monitored object and calculate the second-order motion trend degree; The judgment module is used to judge whether the second-order motion trend degree is the same as the motion trend degree. If they are not the same, the motion trend degree is updated to the second-order motion trend degree.

5. The intelligent fence monitoring and management system for animal husbandry according to claim 1, characterized in that, The early warning sending module includes: The logic module is used to generate an early warning message when the motion trend degree is greater than the preset judgment threshold; The response module is used to respond to the generation result of the early warning message and send the early warning message to the terminal.

6. The intelligent fence monitoring and management system for animal husbandry according to claim 5, characterized in that, The preset judgment threshold is 0.

8.

7. The intelligent fence monitoring and management system for animal husbandry according to claim 1, wherein The speed is collected by an acceleration sensor worn on the monitored object.

8. The intelligent fence monitoring and management system for animal husbandry according to claim 1, characterized in that The motion direction is collected by a gyroscope worn on the monitored object.

Images